Human pancreatic cancers are hypoxic tumors that respond poorly to existing chemotherapy agents. These tumors have adapted to grow in adverse conditions of cellular stress, notably hypoxia and an excess of reactive oxygen species, (ROS). Imexon is an aziridine-containing small molecule which binds to biological thiols and induces oxidative activation of apoptosis in human multiple myeloma cells. In human pancreatic cancer cells imexon induces similar oxidative effects and demonstrates significant synergy with gemcitabine in vitro and in vivo, including regressing established Panc-1 tumors in SCID mice. Mechanistic studies have shown that imexon causes accumulation of ROS in human pancreatic cancer cells, rendering cells unable to progress through G2 phase of division. Further studies have shown that imexon inhibits HIF-1a expression and remains active in hypoxic pancreatic cancer cells, which are shown to have an excess of (O2.-) due to hypoxia-induced inhibition of electron transport at mitochondrial complex III. These results suggest that imexon and related cyanoaziridine-based analogs produce selective cytotoxicity in hypoxic pancreatic cancer cells which already have an excess of ROS. There are three specific aims proposed in this application: (1) evaluate the role of oxidative stress, hypoxia and HIF-1alpha in hypoxic and normoxic human pancreatic cancer cells, (2) evaluate mechanistic effects of pro-oxidant cyanoaziridines as selective therapeutic agents for hypoxic pancreatic cancer cells, and (3) conduct a Phase II clinical trial of imexon in patients with refractory pancreatic cancer, using plasma thiols and RT-PCR-based gene expression studies as potential biomarkers for imexon's oxidative mechanism of action. Three cyanoaziridine analogs will be evaluated pre- clinically as second-generation agents with greater cytotoxic potency over imexon. This project will interact with Project 1 (HIF-1a and hypoxic response), by evaluating the HIF-1a-active compounds in pancreatic cancer models, and by collaborating on the hypoxic mechanisms of the cyanoaziridines using techniques developed in Project 1. Interactions with Project 3 (developmental imaging), will involve the development and validation of imaging techniques to detect oxidation in pancreatic tumors in vivo with eventual extension to clinical trials.